Geek asks and answers

The sequence of different increments with R?

I am optimizing several different models, one of which is radialSVM using the package caret. I am creating a tuning grid in preparation for the loop cycle to find the best parameters used for the model.

One thing that would be extremely useful is some sort of variable increment sequence. For example, I would like to start with small parameter values ​​increasing with small steps. The more I walk, the more steps I can take. I found that small parameters change the model a bit, so I would like to study them more carefully.

It would be fantastic to have the sequence increment by some factor of the current step, say x <- x+5*x. Is this possible with something that already exists (e.g. creative use seq()), or do I need to use a loop?

+5
source share
2 answers

How about something like this:

0.0001 * 6^(0:10)
#  [1]    0.0001    0.0006    0.0036    0.0216    0.1296    0.7776    4.6656
#  [8]   27.9936  167.9616 1007.7696 6046.6176
+5
source

You can use exponential distribution:

qexp((1:100)/100)

> qexp((1:100)/100)
  [1] 0.01005034 0.02020271 0.03045921 0.04082199 0.05129329 0.06187540 0.07257069 0.08338161
  [9] 0.09431068 0.10536052 0.11653382 0.12783337 0.13926207 0.15082289 0.16251893 0.17435339

Adjust it to suit your needs:

 diff( 20* qexp((1:100)/100) )
 [1]  0.2030474  0.2051300  0.2072557  0.2094260  0.2116422  0.2139058  0.2162183  0.2185814
 [9]  0.2209967  0.2234660  0.2259911  0.2285739  0.2312164  0.2339208  0.2366892  0.2395238
+2
source

More articles:


All Articles